Objective: To assess the reliability of the histological diagnosis of bladder cancer by assessing the interobserver variability of staging and grading in pTa/pT1 tumours and evaluating the clinical significance of discrepancies.
Materials and methods: All sections from 301 superficial bladder carcinomas were reviewed by one pathologist. The prognostic relevance of grade and stage from both the initial and review diagnosis were determined in 128 patients for whom there was long-term follow-up information.
Results: There were significant interobserver differences in both the grading and staging of tumours. From a total of 235 tumours that were initially considered pT1, the reviewer classified 35% as pTa, 56% as pT1, 6% as pT1- (at least pT1), and 3% as pT2-4. In 39% of all biopsies there were interobserver differences in tumour grade. The prognostic significance of grade and stage differed between the initial pathology report and the reviewer's diagnosis. The reviewer's staging allowed a better estimate of the risk of subsequent tumour progression than the initial staging. Progression was significantly more common in 49 tumours in which the reviewer agreed with stage pT1 than in 29 tumours that were down-staged from pT1 to pTa (P = 0.0116). However, the initial tumour grade (P = 0.0386) but not the reviewer's grade (P = 0.2645) was significantly linked to progression.
Conclusions: These results show that grading and staging by different pathologists have varying prognostic implications. If possible, biopsies from bladder tumours should be independently evaluated by two different pathologists before radical therapy is administered.